Kevin Knight

Multi-Source Neural Translation

2016

We build a multi-source machine translation model and train it to maximize the probability of a target English string given French and German sources. Using the neural encoderdecoder framework, we explore several combination methods and …

Leveraging Entity Linking and Related Language Projection to Improve Name Transliteration

2016

Traditional name transliteration methods largely ignore source context information and inter-dependency among entities for entity disambiguation. We propose a novel approach to leverage state-of-the-art Entity Linking (EL) techniques to automatically correct name transliteration results, using …

Extracting Structured Scholarly Information from the Machine Translation Literature

2016

Understanding the experimental results of a scientific paper is crucial to understanding its contribution and to comparing it with related work.We introduce a structured, queryable representation for experimental results and a baseline system that automatically …

Deciphering Related Languages

2017

We present a method for translating texts between close language pairs. The method does not require parallel data, and it does not require the languages to be written in the same script. We show results …

AMR Beyond the Sentence: the Multi-sentence AMR corpus

2018 · International Conference on Computational Linguistics

There are few corpora that endeavor to represent the semantic content of entire documents. We present a corpus that accomplishes one way of capturing document level semantics, by annotating coreference and similar phenomena (bridging and …

Two Approaches to Building Collaborative, Task-Oriented Dialog Agents through Self-Play

2021 · arXiv (Cornell University)

Task-oriented dialog systems are often trained on human/human dialogs, such as collected from Wizard-of-Oz interfaces. However, human/human corpora are frequently too small for supervised training to be effective. This paper investigates two approaches to training …

Transfer Learning for Low-Resource Neural Machine Translation

2016 · Digital Collections portal (Koç University)

The encoder-decoder framework for neural machine translation (NMT) has been shown effective in large data scenarios, but is much less effective for low-resource languages. We present a transfer learning method that significantly improves BLEU scores …

Generating Topical Poetry

2016

We describe Hafez, a program that generates any number of distinct poems on a usersupplied topic. Poems obey rhythmic and rhyme constraints. We describe the poetrygeneration algorithm, give experimental data concerning its parameters, and show …

Cross-lingual Name Tagging and Linking for 282 Languages

2017

The ambitious goal of this work is to develop a cross-lingual name tagging and linking framework for 282 languages that exist in Wikipedia. Given a document in any of these languages, our framework is able …

Plan-and-Write: Towards Better Automatic Storytelling

2019 · Proceedings of the AAAI Conference on Artificial Intelligence

Automatic storytelling is challenging since it requires generating long, coherent natural language to describes a sensible sequence of events. Despite considerable efforts on automatic story generation in the past, prior work either is restricted in …

Unsupervised Neural Hidden Markov Models

2016

In this work, we present the first results for neuralizing an Unsupervised Hidden Markov Model. We evaluate our approach on tag induction. Our approach outperforms existing generative models and is competitive with the state-of-the-art though …